2020 · 14 citations · 8 references
ComputingComputer ScienceElectrical EngineeringArtificial IntelligenceMachine LearningPredictive AnalyticsDecision TreesPower Grid AnalysisPower QualityPower System StabilityPower SystemsSupport Vector MachinesElectrical Power SystemsGaussian Process RegressionStability Margin AnalysisStability PredictionSupport Vector RegressionVoltageVoltage Stability
In recent years, machine learning regression methods have attracted a lot of interest to evaluate voltage stability margin in power systems for online application. This paper compares the voltage stability margin prediction using different popular machine learning algorithms for normal operating conditions as well as for contingency cases. Further, this paper also compares the predictive capability of different machine learning algorithms to unseen network topologies. The algorithms used in this study are Gaussian Process Regression, Artificial Neural Network, Support Vector Machine, and Decision Tree. The voltage magnitude and angle at each bus serve as inputs to machine learning algorithms for evaluation of the voltage stability margin. All the analysis was done in the New England 39-bus system and the results show that the Gaussian process regression method provides the predicted model with the highest accuracy for most cases.
8
Pankaj D. Achlerkar, Subhransu Ranjan Samantaray, M. Sabarimalai Manikandan · IEEE Transactions on Smart Grid · 2016 · 276 citations
Voltage Stability Prediction Using Active Machine Learning
Vuk Malbaša, Ce Zheng, Po‐Chen Chen et al. · IEEE Transactions on Smart Grid · 2017 · 143 citations